Rat with Dual Opamp – mind helping with the last section?

[T-Dub]

Hello!
I'm mucking around with a circuit and would love some knowledge dropped on me.
I'm wanting to make a Dual Opamp Rat but the second stage is bugging me.

Here are my questions:
– Original Rat inverts the signal – is there a reason I shouldn't keep both IC's in a non-inverting state? I'm wanting to eventually wrap a clean blend around it so I assume I should keep both IC's the same
– Will I have enough output to just run the second ic as a buffer?
– If I don't, and want to add some gain, I'm getting confused when to send something to VR vs when to send something to ground.
– I'm also a bit confused as to when to add a VR reference at the input of the IC – some circuits use it (like a HoneyBee), and others don't (like a Timmy)

Here's my schematic

[patricks]

Hello!
I can help a bit, I've been breadboarding a Rat-related circuit and the forum members have been super helpful
Some answers below.

Here are my questions:
– Original Rat inverts the signal – is there a reason I shouldn't keep both IC's in a non-inverting state? I'm wanting to eventually wrap a clean blend around it so I assume I should keep both IC's the same
The original Rat doesn't invert the signal, it uses the non-inverting input to the op-amp. The inverting input also has the feedback network applied there, and the high pass RC filtering (https://www.electrosmash.com/proco-rat)

– Will I have enough output to just run the second ic as a buffer?
I don't think the buffer will mind much if it's being fed with a low output signal. If you want some gain from that stage (for recovery after clipping and the tone control) experiment with the value of R11

– If I don't, and want to add some gain, I'm getting confused when to send something to VR vs when to send something to ground.
Vref would be applied at R9 rather than C14

– I'm also a bit confused as to when to add a VR reference at the input of the IC – some circuits use it (like a HoneyBee), and others don't (like a Timmy)
The Timmy uses a Vref, applied to the inverting input: http://revolutiondeux.blogspot.com/2008/05/wanna-build-timmy-overdrive.html

[T-Dub]

Hello!
I can help a bit, I've been breadboarding a Rat-related circuit and the forum members have been super helpful
Some answers below.

Here are my questions:
– Original Rat inverts the signal – is there a reason I shouldn't keep both IC's in a non-inverting state? I'm wanting to eventually wrap a clean blend around it so I assume I should keep both IC's the same
The original Rat doesn't invert the signal, it uses the non-inverting input to the op-amp. The inverting input also has the feedback network applied there, and the high pass RC filtering (https://www.electrosmash.com/proco-rat)

– Will I have enough output to just run the second ic as a buffer?
I don't think the buffer will mind much if it's being fed with a low output signal. If you want some gain from that stage (for recovery after clipping and the tone control) experiment with the value of R11

– If I don't, and want to add some gain, I'm getting confused when to send something to VR vs when to send something to ground.
Vref would be applied at R9 rather than C14

– I'm also a bit confused as to when to add a VR reference at the input of the IC – some circuits use it (like a HoneyBee), and others don't (like a Timmy)
The Timmy uses a Vref, applied to the inverting input: http://revolutiondeux.blogspot.com/2008/05/wanna-build-timmy-overdrive.html

Oh awesome! You've pointed out some brainfarts, so thank you!
If I could borrow your eyeballs and brain a little longer....


This was how I was going to buffer...


and this was how I was going to add a bit of gain – I think I can send it to ground (instead of vref) because it's not on the input? Do I gain anything by sending it to ground to vs vref

[patricks]

Hello!
I can help a bit, I've been breadboarding a Rat-related circuit and the forum members have been super helpful
Some answers below.

Here are my questions:
– Original Rat inverts the signal – is there a reason I shouldn't keep both IC's in a non-inverting state? I'm wanting to eventually wrap a clean blend around it so I assume I should keep both IC's the same
The original Rat doesn't invert the signal, it uses the non-inverting input to the op-amp. The inverting input also has the feedback network applied there, and the high pass RC filtering (https://www.electrosmash.com/proco-rat)

– Will I have enough output to just run the second ic as a buffer?
I don't think the buffer will mind much if it's being fed with a low output signal. If you want some gain from that stage (for recovery after clipping and the tone control) experiment with the value of R11

– If I don't, and want to add some gain, I'm getting confused when to send something to VR vs when to send something to ground.
Vref would be applied at R9 rather than C14

– I'm also a bit confused as to when to add a VR reference at the input of the IC – some circuits use it (like a HoneyBee), and others don't (like a Timmy)
The Timmy uses a Vref, applied to the inverting input: http://revolutiondeux.blogspot.com/2008/05/wanna-build-timmy-overdrive.html

Oh awesome! You've pointed out some brainfarts, so thank you!
If I could borrow your eyeballs and brain a little longer....


This was how I was going to buffer...


and this was how I was going to add a bit of gain – I think I can send it to ground (instead of vref) because it's not on the input? Do I gain anything by sending it to ground to vs vref

You're welcome! I've been helped out so much on the forum with many brain farts, I'm glad I can pay it forward
The buffer option looks good to me (but I'll defer to larger brains if I've missed anything).
The gain option looks good, too. My understanding is that you want to reference that feedback network to GND rather than Vref to keep it as "clean" as possible. Have a read through this thread where Antonis helped me puzzle through the very same thing, also with a Rat schematic I'm modifying (I'm up to the stage of having prototype PCBs in hand but haven't built it up yet. So close!)

[GGBB]

R8 and R10 are unnecessary. There's no harm in R8 except that it robs about 10% of your output. Without R10, POT1 (volume) can be smaller e.g. 10k (R10 and POT1 in parallel form a single resistor). You probably don't need to add gain to the output buffer, but if you do, R11 and R12 should be bigger like 100k and should go to ground through C15 which should be ~220n. If the plan is to tame high end with C13, it does very little because gain is only 2. But making it significantly larger e.g. 10-100n will reduce most other frequencies as well (not dramatically because gain is low). As you raise the gain of this stage, C13 will have a bigger effect and will have less effect on lower frequencies.

[antonis]

There's no harm in R8 except that it robs about 10% of your output.

It could be of somewhat lower value (down to 100R say) for oscillation prevention in case of shorted (or connected to a vely low impedance) outpout..

[ElectricDruid]

This was how I was going to buffer...

This looks fine to me.

and this was how I was going to add a bit of gain – I think I can send it to ground (instead of vref) because it's not on the input? Do I gain anything by sending it to ground to vs vref

No, you have to send it to Vref, but it's nothing to do with it "not being on the input". You have to think about the DC levels. Ignore the audio for a moment and just think about where the voltages sit.

The buffer is fine because its +ve input is fixed at Vref by R9/1M. The audio input and output are capacitively coupled, so no DC from them can interfere. So the op-amp sits happily at a mid-point voltage waiting for audio to come along.

The stage with gain has the same input, so that's good and sets a DC level. The output will therefore go to that same level. However, the foot of the feedback network is at ground, so there's a DC potential (of about 1/4 of the supply) being fed back to the -ve input by the divider R11/R12. That puts the -ve input at a different potential to the +ve input, and op-amps hate that! They *always* try to have their inputs at the same level. If you take the bottom of the divider to Vref, there's no DC potential across it and the problem goes away.
However, there is another better solution which you already have on your schematic, which is to add a cap at the bottom of the feedback divider (like C6 or C7 on the first stage). No DC can cross the cap, so the divider becomes AC-only and will provide feedback for audio but not affect the DC level. Another way of stating this same thing is to say that the cap forms a highpass filter and the gain rolls off to zero at DC.

[antonis]

However, the foot of the feedback network is at ground, so there's a DC potential (of about 1/4 of the supply) being fed back to the -ve input by the divider R11/R12. That puts the -ve input at a different potential to the +ve input, and op-amps hate that!

You could just tell that there is also DC gain (1+R11/R13), Tom..

Sorry, didn't read the following..

Another way of stating this same thing is to say that the cap forms a highpass filter and the gain rolls off to zero at DC.

[antonis]

The Timmy uses a Vref, applied to the inverting input: http://revolutiondeux.blogspot.com/2008/05/wanna-build-timmy-overdrive.html

Not really...

2^nd op-amp is directly DC biased at Vref by 1^st stage output (no DC blocking cap..)
3k3 resistor Vref connection is another cap saver common minor(*) design flaw..


(*) In an ideal world, ideal op-amps shouldn't mind at all but in a real world op-amps suffer both from input bias currents (especially bi-polar ones) and input offset voltages..

[patricks]

Ah, cool, thank you
For the OP this is another great point for reading up on, RG had a good article at Geofex about op amp biasing and making sure differential voltages at the inputs don't get too wide

[287m]

Dylan159 and Kanengomibako did dual opamp RAT.

[T-Dub]

Wow, honestly I am chuffed.
I now have my underlying confusion cleared up around vref vs ground, and how the cap blocks the dc so therefore can go to ground.

Even got two examples with the two different implementations.

Thank you all so much

[antonis]

Dylan159 did dual opamp RAT.

A decent design with IN-OUT in phase and no Vref connections puzzlement..
(with a minor objection about R6/R7 values..)

[Elektrojänis]

There's no harm in R8 except that it robs about 10% of your output.

It could be of somewhat lower value (down to 100R say) for oscillation prevention in case of shorted (or connected to a vely low impedance) outpout..

Or preventing oscillation beacause of cable capacitance... It only aplies when volume pot is maxed though. When it's not the upper half of the vol pot will do the same.

If you want to prevent the oscillation with minimal "signal robbing", you can put the resistor between the pot wiper and output. 10% is under 1dB though, so it should not make much difference.

[antonis]

If you want to prevent the oscillation with minimal "signal robbing", you can put the resistor between the pot wiper and output.

In such a case, when Vol pot is turned all the way down (dead silent output) op-amp has to fight with DC blocking cap..
(it's not good to have a severely oscillating amp even if you don't hear anything, is it..??)

For unknow or dynamic load, it should be better to place that resistor inside the NFB loop..
(with just feedback cap one leg rewiring..)

[Elektrojänis]

If you want to prevent the oscillation with minimal "signal robbing", you can put the resistor between the pot wiper and output.

In such a case, when Vol pot is turned all the way down (dead silent output) op-amp has to fight with DC blocking cap..
(it's not good to have a severely oscillating amp even if you don't hear anything, is it..??)

I probably didn't express clearly enough what I meant, but maybe a Q&D mockup pic based on OP's design will explain. This is what I meant:



I dont see how turning the volume pot down would make the opamp fight with the output coupling cap. There will be full 100kOhm of the pot between the output cap and ground.

For unknow or dynamic load, it should be better to place that resistor inside the NFB loop..
(with just feedback cap one leg rewiring..)

That one probably goes beyond my understanding, but to me it seems like placing the resistor inside the feedback loop partially cancels what I wanted to do. I wanted to separate the cable capacitance connected to the output of the pedal from the output of the opamp and the feedback arrangement. The cable capacitance (you could think of it as C_L in that image) could actually stop the feedback for high frequencies as it will form a low pass filter with R_X. C_F should mitigate that if it's big enough though.

Anyway.. .I don't really see a need for keeping output impedance under 1kOhm when we are talking about guitar pedals. Using 100kOhm volume pot on the output will make the output impedance higher anyway if it's not turned up to max.

[antonis]

I probably didn't express clearly enough what I meant, but maybe a Q&D mockup pic based on OP's design will explain. This is what I meant:

My bad, then..
(mistakenly imagined configuration..)

[T-Dub]

OK gang, think I've got my head around it - does this clean blend make sense?

[T-Dub]

...and one more question – will a simple low pass on the clean work/be enough?

[antonis]

It makes sense but only for Vol pot FCW setting..

For Vol & Blend pots setting at 50% say, you don't have Dry-Wet "equal" mixing..

Better move Vol pot in place of R14 (make it 100k or so), make Blend pot 10k(*) and add 1k or so series resistors on both IC1.A & IC2.B outputs..

(*) For more effective mixing, retain Blend pot 100k value..